PROJECT SUMMARY This proposal is submitted in response to the NIH Development of Animal Models and Related Biological Materials for Research (R21) program. The proposal develops an image-guided robotic platform that performs the automated delivery of molecular genetic tools and non-genetically encoded reagents such as chemical libraries, fluorescent dyes to monitor cellular processes, functionalized magnetic beads, or nanoparticles into thousands of Drosophila embryos in a single experimental session. The proposed work builds on recent engineering innovations in our collaborative group which has developed image-guided robotic systems that can precisely interface with single cells in intact tissue. The two Specific Aims provide for a systematic development of the proposed technologies. AIM 1 first engineers a robotic platform (?Autoinjector?) that can scan and image Drosophila embryos in arrays of egg laying plates. We will utilize machine learning algorithms for automated detection of embryos, followed by thresholding and morphology analysis to detect embryo centroids and annotate injection sites. In AIM 2, we will utilize microprocessor-controlled fluidic circuits for programmatic delivery of femtoliter to nanoliter volumes of reagents into individual embryos. We will quantify the efficacy of the Autoinjector by comparing the survival, fertility, and transformation rates of transposon or PhiC31-mediated transgenesis to manual microinjection datasets. Finally, we will demonstrate the efficient delivery of sgRNAs and mutagenesis in the presence of Cas9. This project fits very well within the goals of the program by engineering a novel tool for producing and improving animal models. The Autoinjector will accelerate Drosophila research and empower scientists to perform novel experiments and genome-scale functional genomics screens that are currently too inefficient or labor intensive to be conducted on a large scale and may additionally enable other novel future applications.